Knowing your way around a probability distribution is a necessity for folks who want to become data scientists and develop cutting-edge applications using the latest AI technologies and techniques. But even the non-unicorns among us can benefit from having a modicum of data literacy, as more of the everyday lives of regular people transition into the digital realm.
It can be hard to grasp the implications of the massive technological and business shifts that are currently underway. Thanks to COVID, digital transformation efforts have been kicked into overdrive, as companies seek ways to not only survive, but thrive in a quickly evolving digital economy. Vast fortunes will be made based on the skill, hard work, and luck of the entrepreneurs employing big data and AI technologies.
AI is being infused into aspects of our lives that we thought would never be touched. In his new Coursera course, AI For Everyone, Andrew Ng says he and his colleagues used to think that hairdressers were probably off-limits to AI–until a hairdresser told him an AI robot could probably replicate Ng’s haircut.
You might think the small business accounting space would be relatively free from AI, but that’s not the case, either. The New Zealand-based software company Xero is actively using AI and machine learning to eliminate the less pleasant aspects of a bookkeeper’s day-to-day life, such as reconciling bank statements, pre-filling expense reports, and warning of low cashflow.
“One of the things we do when we build data products at Xero is we try and remove toil,” says Kendra Vant, the company’s executive GM of data. “We’re automating away the parts of the bookkeeper’s role and a small business owner’s role that they’re not particularly enamored with, so they’re like, sweet, I can do this faster and get on to the things that I actually prefer to do.”
Xero practices augmented intelligence with its products, according to Vant, who says the company is just getting started with AI. When the company uses machine learning in its software to suggest a recommendation, it leaves the final decision up to the human.
“The machine learning part of the products we bult exclusively to date are augmenting the human workflow and still asking the human to say, would you like to confirm that this is correct?” she says. “As the human, you should make the absolutely final decision whether or not this is correct.”
Xero, which has more than 3 million subscribers around the world and had revenues last year in excess of 1 billion New Zealand Dollars (about $630 million at current exchange rates), employs its share of data scientists and machine learning engineers to build augmented intelligence into its software. There is plenty of technological and data complexity going on under the covers. But none of that messiness gets exposed to the users.
“The products we build don’t require the folks who use them to be data literate,” Vant says. “I think that’s super important because most small owners aren’t particularly interested in data processing. They’re more interested in being a florist or a plumber or a general contractor.”
While Xero strives to make its AI simple and intuitive, Vant is still adamant that people in all walks of life should be educating themselves on basic data concepts. Gaining a foundational level of familiarity with data will help them navigate this increasingly complex, data- and AI-driven world.
“I think that’s one of the saddest things. ‘Data is not important to me because I’m not a data scientist,’” she tells Datanami. “Learning R is not important to you if you’re not a data scientist. Learning TensorFlow is not important to you if you’re not a data scientist. But understanding when somebody is lying to you with statistics is important if you’re a member of a modern democracy, and many of us have the pleasure of being a member of a modern democracy, so I think data literacy is extraordinarily important to your average Joe.”
Few people were literate 200 years ago, but it didn’t make a huge difference in most people’s lives, because their day-to-day routines were not dependent on being able to read letters, Vant says. That is definitely not the case today, and looking to the future, having a basic understanding of data will be more important than it is today.
“Maybe 20 years ago, when there wasn’t a whole lot of data accessible for us all to process and make use of as we went through our general lives, no it didn’t matter too much if you weren’t a savvy consumer,” she says. “But that’s the not the world we live in, and it’s certainly not the world that our children live in. They’re going to live in a world with an enormous amount of data–some of it is really quite accessible.
“And if you’re not a person who is confident in looking at that data, looking at how it’s being presented to you and saying either, yes I agree with that presentation, or hang on, there’s something a little fishy in there, I think that’s going to become enormously important in being a confident, comfortable human in the 21st century,” she says.
Data literacy will not only benefit people in their work life, but also help them in their consumer lives, Vant says. Being comfortable with statistics and numbers will benefit even the average consumer because they’ll be able to look at things holistically, spot trends, and identify anomalies, she says.
A groundswell of data literacy among the masses could also benefit companies with big data and AI ambitions. That’s because regulations like CCPA and GDPR have made it more difficult for companies to collect data from consumers, which is forcing companies to get both more creative and more upfront with their data collection methods.
In this brave new data world where consumers hold the cards, data literacy will also pay dividends, as consumers increasingly flock to the companies that show they respect their privacy, which may make them more willing to give consent to data collection.
“If I was giving advice to a small business owner…I’d say think really hard when you first ask your customers for a piece of data. Do I actually need it? And would my customers think there’s a reasonable reason for me to have it?” Vant says. “I think there’s a shift with people being more savvy about why does that company want that piece of information, and actually voting with their feet to use businesses, big and small, that are demonstrating being careful with their information.”
All of these factors work together in a way that makes data literacy more important than ever, for both the companies that are employing AI technologies and techniques, as well as the consumers who are enjoying the fruits of the data and AI labors, but don’t want to get caught up in the darker side of the digital economy or fall behind.
“Being data literate is to some extent a really strong overlay between being comfortable in a digital environment,” Vant says. “If you want to sell your products digitally, then you want to be able to make good choices about partners you use and ways that you engage with customers. They’re intertwined in ways that people haven’t realized yet.”
Related Items:
The Death of Third-Party Cookies and the Rise of Zero-Party Data
A Culture Shift on Data Privacy
Testing Data Literacy on Main Street
This article originally appeared in Datanami.